rheology-fit 0.1.0

Fit non-Newtonian flow curve models (Power-law, Herschel-Bulkley, Casson) to shear stress / shear rate data for food rheology.

rheology-fit

Fit non-Newtonian flow curve models (Power-law, Herschel-Bulkley, Casson) to shear stress / shear rate data, with objective model selection and flow-behavior classification — for sauces, purées, doughs, and other non-Newtonian food fluids.

Problem

Rheological characterization of non-Newtonian food fluids — the relationship between shear stress and shear rate — underpins process design (pumping, mixing, extrusion), texture/mouthfeel prediction, and quality control. Researchers typically measure a flow curve on a rheometer and then fit one or more models (Power-law, Herschel-Bulkley, Casson) by hand in a spreadsheet — a slow, inconsistent process, especially since Herschel-Bulkley requires genuine nonlinear regression that spreadsheets handle poorly. rheology-fit automates model fitting, objective model selection (AIC), and flow-behavior interpretation (shear-thinning/thickening, yield stress), producing a publication-ready report.

Installation

pip install rheology-fit

Or from source:

git clone https://github.com/karenkhachatryan-lab/rheology-fit.git
cd rheology-fit
pip install -e ".[dev]"

Usage

Input CSV with columns shear_rate (s⁻¹), shear_stress (Pa), and optionally shear_stress_std:

shear_rate,shear_stress,shear_stress_std
1,20.6,0.3
10,33.2,0.4
50,47.4,0.5
100,55.6,0.6

Fit models and generate a report:

rheology-fit fit data.csv --output report

This produces:

• report.pdf — flow curve with all fitted model curves, flow-behavior annotation, residuals plot for the best model,
• report.json — fitted parameters, metrics (R², RMSE, AIC), and flow-behavior classification.

Print citation information:

rheology-fit cite

Models

• Power-law (Ostwald-de Waele): τ = K·γ̇ⁿ — 2 parameters, no yield stress.
• Herschel-Bulkley: τ = τ₀ + K·γ̇ⁿ — 3 parameters; the standard model for yield-stress food fluids (ketchup, mayonnaise, purées).
• Casson: √τ = √τ₀ + √(η꜀·γ̇) — 2 parameters; common for chocolate and some dairy/meat emulsions.

Unlike isotherm-fit's BET model, all three models here are fit on the full dataset, so their AIC values are always directly comparable — no restricted-range caveat.

Project status

Early development (v0.1.0) — MVP scope: 3 models, CSV loader, AIC-based model selection, PDF/PNG + JSON report generation, CLI via Typer. See CHANGELOG.md.

Citing this software

If you use rheology-fit in your research, please cite it — see CITATION.cff or run rheology-fit cite for the formatted citation and BibTeX entry. DOI: 10.5281/zenodo.21713809.

Contributing and support

Bug reports, feature requests, and usage questions are all welcome via GitHub Issues.

License

MIT — see LICENSE.